iblai-vibe-local-llmlisted

Use when adding on-device LLM inference (Ollama backend) to a vibe Next.js + Tauri app. Documents the contract a vibe app must implement so the SDK can route chat through a local model via `localLLMProps` — the Tauri command names, event names, and the React hook shape. Mention before wiring "offline AI", "local model", "on-device LLM", or "Ollama" support into a desktop build.
iblai/vibe · ★ 15 · AI & Automation · score 76

Install: claude install-skill iblai/vibe

# /iblai-vibe-local-llm A vibe Next.js + Tauri app can serve chat from a model running on the user's machine instead of the hosted backend. The SDK turns this on automatically when you pass a `localLLMProps` object whose `isAvailable` is `true` and whose `ollamaStatus.model_installed` (or `foundryStatus.has_models`) reports a ready model. This skill defines the **contract** your app must implement to feed that prop: - the Tauri `#[command]` names the React layer calls, - the Tauri event names the React layer listens on, - the shape the React hook must return, and points to references describing the install-flow, the on-disk layout, and the failure modes you'll hit. It does **not** ship a reference implementation — local-LLM support is closed-source in the apps that have it today. > Do NOT add custom styles / colors to ibl.ai SDK components. They > ship with their own styling. > See [BRAND.md](https://raw.githubusercontent.com/iblai/vibe/refs/heads/main/BRAND.md) > for the brand palette. > **Common setup (env, conventions, verification):** see > [docs/skill-setup.md](https://raw.githubusercontent.com/iblai/vibe/refs/heads/main/docs/skill-setup.md). ## Prerequisites 1. Tauri shell wired up via [`/iblai-vibe-ops-build`](../iblai-vibe-ops-build/SKILL.md) (adds `src-tauri/` + `pnpm install`). The skill assumes `src-tauri/` already exists. **Heads-up:** on an app that already has a non-trivial `next.config.mjs`, adding the Tauri shell can mis-merge its `output: